The ability to detect the presence and composition of chemical species has been an important goal for several reasons. For example, the detrimental environmental effects of toxic species such as formaldehyde, carbon monoxide, ozone, hydrocarbons, chlorocarbons, nitrogen oxides, aromatics and heavy metals has led to the need to develop efficient, sensitive, and affordable ways of detecting the composition and presence of such toxic substances. Additionally, the efficiency of chemical processes, in terms of energy and raw material used per unit product or service delivered, relies on the ability of the overall system to reliably sense deviations from the optimal processing conditions. Since process efficiencies directly determine the overall costs of the process and indirectly determine the wastes generated by the process, it is critically important that a method be available that can provide the necessary feedback about the process (sensors) and initiate actions to evolve the system parameters to the optimal levels (actuators).
The temperature, pressure and flow monitoring and control of chemical, environmental, biochemical, biomedical, geological, metallurgical, and physical processes have been extensively researched and the state-of-the-art technologies quite effectively enable real-time evolution of the monitored process. However, compositional monitoring and control of these processes leaves much to be desired. Crude methods for process monitoring and control are based upon batch analysis, i.e., a statistical set of samples are obtained ("grab sample" approach) and then analyzed for composition. These data are then interpreted and actions are initiated to control the process to desired levels. The response time for such a strategy often is in days, if not weeks. This strategy has serious deficiencies since it inherently accepts inefficient operation between the time the samples were obtained and the actions are initiated to correct deviations from the optimal. Yet another deficiency in such a strategy is that it overlooks the possibility that the process conditions may have changed during the response and analysis lag time.
Alternatively, sophisticated monitors (such as gas chromatographs with suitable sampling and transport systems) have been integrated into the processes. These systems are expensive, bulky, not suited for extreme temperatures and pressures, and have response times that are more than a few minutes. Real-time composition monitoring and control of the chemical and combustion processes require sensors that overcome these limitations. Specifically, sensor technology for gas sensing applications should ideally be selective, sensitive to trace species, fast (short response time), small, accurate, reproducible, stable in extreme environments, durable (long life), and affordable.
Prior art methods for producing gas sensors include those of U.S. Pat. No. 4,631,952 which teaches a method of preparing a sensor by the formation of a dispersion of conducting particles with a material capable of swelling in the presence of the liquid, gas or vapor to be sensed. Furthermore, U.S. Pat. No. 4,984,446 teaches the preparation of a gas detecting device by a layer by layer build up process, and U.S. Pat. No. 4,988,539 teaches an evaporation-deposition method and process for manufacturing a gas detection sensor. Finally, U.S. Pat. No. 5,387,462 teaches a method of making a sensor for gas, vapor, and liquid from a composite article with an electrically conductive surface having an array of whisker-like microstructures with an aspect ratio of 3:1 to 100:1.
Although these prior methods provide improved methods for producing sensors, there is still a need to develop sensors which are selective, sensitive to trace species, fast, small, accurate, reproducible, stable in extreme environments, durable, and finally affordable.